Read on Twitter
A just-out report again calls for a new digital regulator. Authors are @tewheels (frm. FCC Chair), Gene Kimmelman (who also worked on Stigler Center report) & Phil Verveer. I've written and am writing on this topic; so wanted to see this report. Here's what I found. THREAD. https://twitter.com/ashleyrgold/status/1296422011091267586
I have a lot of Qs about the *content* of the proposed new regulation. (e.g., The report suggests it is adopting a common law-inspired approach. I'm a huge fan of CL approaches. But the proposal looks nothing like common law - & actually abandons the best features of common law.)
But for this thread I'll focus on the *form* of regulator: Why a new agency? And what form would that agency take?
Why a new agency? Current agencies are already stretched too thin, it says. But Congress could just add resources to a current agency instead of creating a new one. The report doesn't address how a new agency would be more cost-effective than strengthening an existing one.
The real reason, it seems, is what I'll dub the "no old fogeys" argument. The authors think current agencies are too "analog" and they want an agency with "digital DNA."
This is a strangely blinkered position. For a report somewhat obsessed with history, it seems unaware of the role the FTC has long played in platform regulation. The report repeatedly describes areas such as "privacy and the security of personal information" as "not protected."
And when the report lists out all the very important things the FTC already does, they somehow leave out privacy and data security enforcement:
Yet the FTC has been bringing cases against platforms for 30+ years (GeoCities!), has hundreds of privacy and data security enforcement actions, and has pretty much every major platform under order for something or other. https://www.ftc.gov/news-events/press-releases/2020/02/ftc-releases-2019-privacy-data-security-update
They do mention antitrust cases, but the consumer protection cases are more numerous and cover many of the areas that apparently require a new agency. The authors want to start over with a blank slate, but it appears they haven't really looked at the current options.
Re: the agency's form: Report encourages a risk-based approach focused on outcomes instead of compliance procedures. As such, it urges that Congress make platform obligations "as general and flexible as circumstances permit." and let the agency fill in with "agile" government.
The report recommends this broad objectives approach to avoid the dated rules-based approach. "[A]t a time when industrial management was a top-down, rules-based bureaucracy, the agencies created to oversee industrial activity adopted a similar approach."
Again, it seems like they totally forgot about the FTC, which has a mandate to address "unfair methods of competition" and "unfair and deceptive acts or practices." Those are very broad objectives that are not at all rule-based! The FTC has a lot of experience using those tools.
The report is very vocal about the need to take a totally new regulatory approach, but the proposed Digital Platform Agency (DPA) is at the core a bipartisan independent APA rulemaking agency with a Code Council that one can think of as "super commenters."
The report emphasizes that this structure will enable agile regulation that can keep up with the pace of regulatory change. But it doesn't do a great job of explaining how APA rulemaking with an extra step will be faster than regular ol' APA rulemaking.
The biggest gap is that the report doesn't explain how this new reg. structure will prevent the single biggest problem with industry-specific regulators: regulatory capture. I've explained why this is so important: https://www.washingtonpost.com/outlook/2019/10/30/creating-new-federal-agency-regulate-big-tech-would-be-disaster/
In summary, the report does offer one of the most detailed proposals for a DPA. But it fails to make a strong case for establishing an entirely new agency. And it ignores the most fundamental question in government institution building: how to prevent capture. /END
